R is designed for advanced statistical computations. Apart from ready-to-use implementations of state-of-the-art algorithms, R’s other great assets are vector and matrix computations. R transformations complement Python and SQL transformations where computations or other operations are too difficult. Common data operations like joining, sorting, and grouping, however, are still easier and faster to do in SQL Transformations.
The R script is running in an isolated Docker environment. The current R version is R 3.5.0.
The Docker container running the R transformation has allocated 8GB of memory and the maximum running time is 6 hours. The container is also limited to the equivalent of 2 Intel Broadwell 2.3 GHz processors.
The R script itself will be compiled to
/data/script.R. To access input and output tables, use relative
out/tables/file.csv), or absolute (
To access downloaded files, use the
/data/in/user/tag path. If you want to dig really deep, have a look
at the full Common Interface Specification. Temporary files can
be written to the
/tmp/ folder. Do not use the
/data/ folder for files you do not wish to exchange with KBC.
The R script to be run within our environment must meet the following requirements:
The R transformation can use any package available on
CRAN. In order for a package and
its dependencies to be automatically loaded and installed, list its name in the package section. Using
for loading is not necessary then.
The latest versions of packages are always installed. Some packages are already installed in the environment (see full list and their dependencies), these packages do not need to be listed in the transformation.
Tables from Storage are imported to the R script from CSV files. The CSV files can be read by standard R functions.
Generally, the table can be read with default R settings. In case R gets confused, use the exact format
sep=",", quote="\"". For example:
Do not use the row index in the output table (
row.names=FALSE). If you are using the
readr package, you can also use the
which doesn’t write row names.
The row index produces a new unnamed column in the CSV file which cannot be imported to Storage.
We have set up our environment to be a little zealous; all warnings are converted to errors and they cause the
transformation to be unsuccessful. If you have a piece of code in your transformation which may emit warnings
and you really want to ignore them, wrap the code in a
We recommend that you create an RStudio sandbox with the same input mapping your transformation will use. This is the fastest way to develop your transformation code.
Tip: Limit the number of rows you read in from the CSV files:
This will help you catch annoying issues without having to process all data.
To simulate the input and output mapping, all you need to do is create the right directories with the right files. The following image shows the directory structure:
The script itself is expected to be in the
data directory; its name is arbitrary. It is possible to use relative directories,
so that you can move the script to a KBC transformation with no changes. To develop a Python transformation which takes
a sample CSV file locally, take the following steps:
in/tablessubdirectory of the working directory.
in/usersubdirectory of the working directory, and make sure that their name is without any extension.
Use this sample script:
A complete example of the above is attached below in data.zip.
Download it and test the script in your local R installation. The
result.csv output file will be created.
This script can be used in your transformations without any modifications.
All you need to do is
source.csv(expected by the R script),
result.csv(produced by the R script) to a new table in your Storage,
It is possible to output informational and debug messages from the R script simply by printing them out. The following R script:
produces the following events in the transformation job:
app$logError functions are internally available; they can be useful if you need to know the precise
server time of when an event occurred. The standard event timestamp in job events is the time when the event was received
converted to the local time-zone.
The above steps are usually sufficient for daily development and debugging of moderately complex R transformations. Although they do not reproduce the transformation execution environment exactly. To create a development environment with the exact same configuration as the transformation environment, use our Docker image.
There are more in-depth examples dealing with